1

Factors Affecting Profitability of Insurance Companies in

Ethiopia: Panel Evidence

Abate Gashaw Ayele

A Thesis Submitted to

The Department of Accounting and Finance

Presented in Partial Fulfillment of the Requirements for the

Degree of Master of Science in Accounting and Finance

Addis Ababa University

Addis Ababa, Ethiopia

May, 2012

0

Signed Declaration

I declare that the thesis for the M.Sc. degree in accounting and finance at the University of

Addis Ababa, herby submitted by me, is my original work and has not previously been

submitted for a degree at this or any other University, and that all references materials

contained therein have been duly acknowledged.

Name Abate Gashaw Advisor‟s Name Dr. P. Laxmikantham

Signature------------------- Signature-----------------------------------

1

Signed Declaration

Addis Ababa University

School of Graduate Studies

This is to certify that the thesis prepared by Abate Gashaw Ayele entitled: Factors Affecting

Profitability of Insurance Companies in Ethiopia: Panel Evidence submitted in partial

fulfillments of the requirements for the Degree of Masters of Science in Accounting and

Finance complies with the rules and regulations of the university and meets the expected

standards with respect of originality and quality. Hence all reference materials contained

therein have been duly acknowledged.

Signed by the Examining Committee

Examiner______________________ Signature______________ Date_______________

Examiner______________________ Signature______________ Date_______________

Advisor _______________________ Signature______________ Date_______________

Chair of Department or Graduate Program Coordinator

iii

Abstract

Factors Affecting Profitability of Insurance Companies in Ethiopia: Panel Evidence

Abate Gashaw

Addis Ababa University, 2012

Profitability is one of the most important objectives of financial management because one

goal of financial management is to maximize the owner` s wealth. This paper examined the

effects of firm specific factors (age of company, size of company, volume of capital, leverage

ratio, liquidity ratio, growth and tangibility of assets) on profitability proxied by ROA.

Profitability is dependent variable while age of company, size of company, volume of capital,

leverage liquidity ratio, growth and tangibility of assets) are independent variables. The

sample in this study includes nine of the listed insurance companies for nine years (2003-

2011). Secondary data obtained from the financial statements (Balance sheet and Profit/Loss

account) of insurance companies, financial publications of NBE are analyzed. From the

regression results; growth, leverage, volume of capital, size, and liquidity are identified as

most important determinant factors of profitability hence growth, size, and volume of capita

are positively related. In contrast, liquidity ratio and leverage ratio are negatively but

significantly related with profitability. Lastly, age of company and tangibility of assets are

not significantly related with profitability.

iv

ACKNOWLEDGEMENTS

I would like to take this opportunity to express my sincere gratitude to Dr. P. Laxmikantham,

department of accounting and finance at Addis Ababa University, for his expert guidance,

helpful criticism, valuable suggestions and encouragement at every stage during the

completion of this work. It was pleasant and inspiring experience for me to work under his

guidance.

I also would like to thank Dr. Venkati Ponnala, department of accounting and finance, Addis

Ababa University, for his initial guidance, valuable suggestions in doing this paper.

It is my pleasure to thank staff members of all insurance companies under study particularly

those staffs working in departments of finance, marketing and corporate planning, who give

me the relevant data that are very much valuable for this study.

I add a special note of admiration and gratitude to my families and friends, without their

moral support, it would have been impossible for me to go through this piece of work.

At last but not least my deep gratitude goes to Debremarkos University (DMU) and Addis

Ababa University (A.A.U) for sponsoring me throughout my work.

v

Table of Contents

Pages

List of Tables .............................................................................................................................. i

List of Figures ............................................................................................................................ ii

List of Appendices .................................................................................................................... iii

List of Abbreviations ............................................ ………………………………………………iv

Chapter one Introduction ........................................................................................................... 1

1.1. Background of the study ................................................................................................... 1

1.2. Background of insurance companies in Ethiopia ............................................................... 2

1.3. Statement of the problem .................................................................................................. 5

1.4. Objectives ......................................................................................................................... 7

1.4.1.General objective ......................................................................................................... 7

1.4.2.Specific objectives ....................................................................................................... 7

1.5. Methods adopted ............................................................................................................... 7

1.6. Limitation and scope of the study ...................................................................................... 8

1.7.Significance and expected outcomes of the study ................................................................ 8

1.8. Structure of the study…... .................................................................................................. 9

Chapter Two Literature Review .................................................................................................10

2.1. Introduction ......................................................................................................................10

2.2. The concept of insurance companies and their financial performance................................11

2.3. The concept of profitability ...............................................................................................12

2.4. Determinates of profitability in insurance companies: an empirical review .......................15

2.4.1. Internal determinants.................................................................................................16

2.4.1.1. Firm size and age .............................................................................................17

2.4.1.2. Liquidity ..........................................................................................................19

2.4.1.3. Leverage ..........................................................................................................19

2.4.1.4. Volume of capital .............................................................................................20

2.4.1.5. Tangibility ........................................................................................................21

2.4.1.6. Growth rate ......................................................................................................22

2.5. Summary of literature .......................................................................................................22

vi

Chapter Three Research Methodology ......................................................................................24

3.1. Introduction ....................................................................................................................24

3.2. Research approach ..........................................................................................................25

3.3. Research method.............................................................................................................25

3.4. Data and data sources .....................................................................................................26

3.5. Sampling mechanism .....................................................................................................28

3.6. Methods of Data Analysis ...............................................................................................28

3.6.1. Descriptive analysis .................................................................................................28

3.6.2. Correlation analysis..................................................................................................29

3.6.3. Regression analysis ..................................................................................................29

3.7. Design of empirical model ...............................................................................................29

3.8. Variable selection and measurement ...............................................................................31

Chapter Four Data Analysis and Findings ..................................................................................37

4.1. Introduction ....................................................................................................................37

4.2. Summary of findings ......................................................................................................54

4.2.1. Age ..........................................................................................................................54

4.2.2. Size ..........................................................................................................................55

4.2.3. Leverage ..................................................................................................................56

4.2.4. Growth .....................................................................................................................56

4,2,5. Tangibility of assets .................................................................................................57

4.2.6. Liquidity ..................................................................................................................57

4.2.7. Volume of capital .....................................................................................................58

Chapter Five: Conclusions and Implications of results ...............................................................61

5.1. Conclusion .....................................................................................................................61

5.2. Implications of the results .............................................................................................61

5.3. Recommendations for future research ............................................................................63

Bibliography

Appendices

i

List of Tables

Table 1.1 List of insurance companies operating in Ethiopia as on 2012 ................................ 5

Table 4.1 Descriptive analysis ...............................................................................................40

Table 4.2 White test regression ............................................................................................42

Table 4.3 Chi Square calculated and tabulated......................................................................42

Table 4.4 Correlations between profitability and independent variables .................................45

Table 4.5 Correlation between age and ROA .........................................................................46

Table 4.6 Correlation between size and ROA .......................................................................46

Table 4.7 Correlation between leverage ratio and ROA .........................................................47

Table 4.8 Correlation between growth and ROA ...................................................................47

Table 4.9 Correlation between volume of capital and ROA ..................................................47

Table 4.10 Correlation between tangibility of assets and ROA ..............................................48

Table 4.11 Correlation between liquidity ratio and ROA ........................................................48

Table 4.12 Collinearity (model 1) ..........................................................................................49

Table 4.13 Collinearity (model 2) ..........................................................................................50

Table 4.14 Model summary (b) ..............................................................................................50

Table 4.15 Panel random effect estimation result after excluding VOC ..................................51

Table 4.16 ANOVA (b).........................................................................................................52

Table 4.17 Collinearity (model 3) random effect regression results excluding size .................53

Table 4.18 Model Summary ...................................................................................................53

ii

List of Figures

Figure 4.1: Histogram-------------------------------------------------------------------------38

Figure 4.2: Normal P-P plot of regression standardized residual-----------------------39

Figure 4.3: Residuals distribution-----------------------------------------------------------44

iii

List of Appendices

Appendix 1 Profitability of selected insurance companies (ROA)

Appendix 2 Hausman Test for panel regression

Appendix 3: Random Effects Regression out put using E-views

Appendix 4: Table designed for collecting raw Panel financial data to be used in regression

analysis.

Appendix 5: Descriptive statistics

Appendix 6: Correlation matrix

Appendix 7: Panel unit root test on ROA

iv

List of Abbreviations

GLS: Generalized Least Square

Lev.: Financial Leverage

NBE: National Bank of Ethiopia

OLS: Ordinary Least Square

ROA: Return on Asset

ROE: Return on Equity

ROIC: Return on Invested Capital

TA: Tangibility of Assets

UAE: United Arab Emirates

UK: United Kingdom

US: United States of America

VOC: Volume of Capital

WACC: Weighted Average Cost of Capital

1

Chapter One: Introduction

1.1 Background of the study

The background of the study deals with the role of financial institutions in the economy of a

country in general and insurance companies in particular and it means their efficient and

effective financial system through savings mobilization, risk transfer and intermediation.

Therefore, financial institutions, channel funds and transfers risks from one economic unit to

another economic units so as to facilitate trade and resources arrangement. Recent research,

as surveyed by Naveed et al (2011), shows that the efficiency of financial intermediation and

transfer of risk can affect economic growth while at the same time institutional insolvencies

can result in systemic crises which have unfavorable consequences for the economy as a

whole. Hence, the important role that financial institutions such as insurance companies

remain in financing and insuring economic activity and contribute to the stability of the

financial system in particular and the stability of the economy of concerned country in

general is part of immune and repair system of the economy

.

According to Hifza Malik (2011) profitability is one of the most important objectives of

financial management since one goal of financial management is to maximize the owners‟

wealth, and, profitability is very important determinant of performance. Therefore, insurance

companies have importance both for businesses and individuals as they channel funds and

indemnify the losses of other sectors in the economy and put them in the same positions as

they were before the occurrence of the loss respectively. In addition, insurance companies

provide economic and social benefits in the society by prevention of losses, reduction in

anxiousness, fear and increasing employment.

2

Therefore, from above expression it can be inferred that, the current business world without

financial institutions such as insurance companies is unsustainable because in one way, it is a

normal practice that some economic units are in surplus while the others remain in deficit

and in the other way risky businesses have not a capacity to retain all types of risk in current

extremely uncertain environment. Performance of financial institutions can affect economic

growth while at the same time institutional insolvencies can result in systemic crises which

have unfavorable consequences for the economy as a whole. Therefore it requires empirical

investigation so as to sort out what are the important factors affecting profitability of

insurance companies and this will help concerned bodies to focus on the relevant factors.

Hence the efficient performance of the institutions has become important and investigations

by different researchers focus on what factors determine the performance especially the

financial performance of the sector.

1.2 Background of insurance company in Ethiopia

For the last decade, the Ethiopian financial institutions in general and insurance companies in

particular have shown the impressive progress in terms of number and service which not only

creates the employment opportunities but also enhances the business activities in the

Ethiopian economy. The work of Hailu Zeleke (2007) explores the historical routes,

examines its emergence and indicates the track that the insurance industry in Ethiopia has

gone through ever since its inception in early twentieth century. It is indicated that there has

hardly ever been any work in insurance business in Ethiopia that went into the historical and

factual aspects of the industry.

3

The history of insurance service is as far back as modern form of banking service in Ethiopia

which was introduced in 1905. At the time, an agreement was reached between Emperor

Menelik II and a representative of the British owned National Bank of Egypt to open a new

bank in Ethiopia. Similarly, modern insurance service, which were introduced in Ethiopia by

foreigners, mark out their origin as far back as 1905 when the bank of Abyssinia began to

transact fire and marine insurance as an agent of a foreign insurance company. According to

a survey made in 1954, there were nine insurance companies that were providing insurance

service in the country. With the exception of Imperial Insurance Company that was

established in 1951, all the remaining of the insurance companies were either branches or

agents of foreign companies. In 1960, the number of insurance companies increased

considerably and reached 33. At that time insurance business like any business undertaking

was classified as trade and was administered by the provisions of the commercial code.

According to Hailu Zeleke (2007), the first significant event that the Ethiopian insurance

market observation was the issuance of proclamation No. 281/1970 and this proclamation

was issued to provide for the control & regulation of insurance business in Ethiopia.

Consequently, it created an insurance council and an insurance controller's office, its strange

impact in the sector. The controller of insurance licensed 15 domestic insurance companies,

36 agents, 7 brokers, 3 actuaries & 11 assessors in accordance with the provisions of the

proclamation immediately in the year after the issuance of the law.

Accordingly as stated by the office mentioned above, the law required an insurer to be a

domestic company whose share capital (fully subscribed) not to be less than Ethiopian Birr

4

400,000 for a general insurance business, Birr 600,000 in the case of long-term insurance

business and Birr 1,000,000 to do both long-term & general insurance business. The

proclamation defined 'domestic company' as a share company having its head office in

Ethiopia and in the case of a company transacting a general insurance business at least 51%

and in the case of a company transacting life insurance business, at least 30% of the paid-up

capital must be held by Ethiopian nationals or national companies.

After four years that is after the enactment of the proclamation, the military government that

came to power in 1974 put an end to all private enterprises. Then all insurance companies

operating were nationalized and from January 1, 1975 onwards the government took over the

ownership and control of these companies & merged them into a single unit called Ethiopian

Insurance Corporation. In the years following nationalization, Ethiopian Insurance

Corporation became the sole operator.

After the change in the political environment in 1991, the proclamation for the licensing and

supervision of insurance business heralded the beginning of a new era. Immediately after the

enactment of the proclamation in the 1994, private insurance companies began to increase.

Currently, there are 14 insurance companies in operation. Both public owned and private

insurance companies which are operating as on January 2012 throughout the country are

listed in the following table 1.1

5

Table 1.1 List of insurance companies operating in Ethiopia as on 2012

S/N Name Type Establishment

Year

1 Ethiopian Insurance Corporation General 1975

2 Africa Insurance company S.C General 1994

3 Awash insurance company S.C General 1994

4 National Insurance company of Ethiopia S.C General 1994

5 Nyala Insurance company S.C General 1995

6 Nile Insurance company S.C General 1995

7 The United Insurance S.C General 1997

8 Global Insurance Company S.C General 1997

9 NIB insurance company General 2002

10 Lion Insurance Company S.C General 2007

11 Ethio-Life Insurance S.C Life 2008

12 Oromia Insurance Company S.C General 2009

13 Abay Insurance Company General 2010

14 Birhan Insurance company S.C General 2011

Source: http://www.nbe.gov.et/financial/insurer.htm accessed January 4, 2012

1.3 Statement of the Problem

The best performance of any industry in general and any firm in particular plays the role of

increasing the market value of that specific firm coupled with the role of leading towards

the growth of the whole industry which ultimately leads to the overall success of the

economy. Measuring the performance of financial institutions has gained the relevance in the

6

corporate finance literature because as intermediaries, these companies in the sector are not

only providing the mechanism of saving money and transferring risk but also helps to

channel funds in an appropriate way from surplus economic units to deficit economic units so

as to support the investment activities in the economy.

The insurance industry in particular is part of immune and repair system of an economy and

successful operation of the industry can set energy for other industries and development of an

economy. To do so the insurance industry is expected to be financially solvent and strong

through being profitable in operation. Hence, not only measuring the financial performance

of insurance companies but also clear insight about factors affecting financial performance in

the industry is then the problem to be investigated. Therefore, the determinants of insurance

company‟s performance have attracted the interest of academicians, practitioners and

institutional supervisors.

The absence of empirical studies in Ethiopia concerning determinants of insurance

companies‟ profitability is then what motivated the researcher to put his own contribution on

what factors affect the financial performance of insurance companies. While taking in to

consideration the absence of empirical inquiry into the factors affecting insurance

companies‟ financial performance, the researcher attempts to work on such untouched

empirical evidence in the country. Hence, these are important issues to be investigated for the

insurance managers, professionals, regulators and policy makers to support the sector in

achieving the excellence so that required economic outcomes could be obtained from the

help of the sector in Ethiopia by understanding the success and failure factors of profitability.

7

In other words, the later problem that is what factors affect the financial performance of

insurance companies has not been adequately investigated. While taking in to consideration

the absence of empirical inquiry into the factors affecting insurance companies‟ financial

performance, the researcher attempts to work on such untouched empirical evidence in the

country.

1.4 Objectives: with regard to the objectives of this study, the researcher tried to address one

broad general objective and some more specific objectives just derived from the general

objective and these are presented below.

1.4.1 General objective

The main objective of the study is to identify and compare the factors determining the

financial performance of the Ethiopian insurance companies for the period of 2003 to 2011.

1.4.2 Specific Objectives

Based on the above general objective, the researcher elucidates the following specific

objectives:

1. To identify the main determinants of insurance companies‟ profitability.

2. To measure the extent to which these determinants exert impact on insurance

companies‟ profitability.

3. To rank the factors according to their degree of influence on insurance companies‟

profitability.

4. To determine the relationship between these factors and profitability in insurance

companies.

1.5. Methods Adopted

In achieving the objectives and obtaining answers for research questions, the study adopted

quantitative method research approach. The method adopted consists of the survey of

8

financial statements of individual insurance companies. With regard to the survey, the target

population consists of nine insurance companies. The number of total insurance companies

under study is nine and observation is also for nine years and then nine times nine, becomes

eighty one total observations included.

1.6 Limitation and scope

Even though there are other formal, semiformal and informal financial institutions, the study

focus only on the determinants of profitability of insurance companies in Ethiopia. As the

researcher tried to point out the scope of the study, the horizon of the study confined merely

on the quantitative measure of determinates of insurance companies profitability (financial

performance) in Ethiopia without any overall performance measurement tool. It would have

also been very useful, if it includes macro-economic factors of profitability. However, due to

the constraints, the researcher is forced to limit the study on this small concern.

1.7 Significance and expected outcome of the study

The main reason for this study is that the researchers have not paid enough attention to this

subject in Ethiopia. Most of the studies previously focused on banks not on insurance

companies as well as some focused on only analysis of financial performance not on factors

affecting financial performance

Therefore, this study is expected to provide empirical evidence on the profitability (financial

performance) of insurance companies in Ethiopia.

Furthermore, many parties would benefit from the results that will emerge from the results of

the study and these parties are:

9

Management: Administration interested in identifying indicators of success and failure to take

the necessary actions to improve the performance of the company and choose

the right decisions.

Government: Government interested in knowing which companies operate successfully or

failed to take the necessary measures to avoid crises of the bankruptcy in these

companies.

Investors: Investors interested in such studies in order to protect their investment, and

directing it to the best investment.

Customers: Customers interested in knowing the ability of insurance companies to pay their

obligations based on the indicators of success of the companies.

This research does have significant role to play in shading light on how to better understand

what determines financial institution‟s such as insurance companies‟ profitability (financial

performance) in Ethiopia. Furthermore, this study does have a paramount importance in

providing a better ground for insurance managers, business professionals, business initiatives

and policy makers. Moreover, the researcher also contributes that this study can potentially

serve as a stepping stone for further research in the area.

1.8 Structure of the study

The reminder of this paper is organized as follows: Chapter two presents the previous studies

by looking at profitability, the factors that determine profitability in insurance companies in

Ethiopia in particular and other financial institutions in general so as to revise relevant

literature. Chapter three presents the research design, methodology and hypotheses

development. Chapter four presents the analysis, findings and results and chapter five

presents the conclusions and implications of the results.

10

Chapter Two: Literature Review

2.1 Introduction

This chapter deals with the concept of insurance, profitability and provides investigations on

factors affecting the profitability or financial performance in insurance companies and also

other financial institutions. Previous researches with regard to profitability mostly focused on

financial institutions. Most of the researches conducted with regard to determinants of banks‟

profitability could be classified in two as country specific such as Belayneh (2011), Tesfaye

(2008), in Ethiopia, Uhomoibhi Toni Aburime (2008), Samy Ben Naceur (2003), where as

others such as Sylwester Kozak (2011), Valentina Flamini, Calvin McDonald, and Liliana

Schumacher (2009) conducted their research at a cross country level.

In these investigations, determinants are classified as internal factors which are under the

control of the management of banks and external factors those are beyond the control of the

management. Therefore, it would be possible to presuppose that oorganizational performance

has attracted scholarly attention in corporate finance literature. However, in the context of

insurance sector, it has received a little attention Hafiz Malik (2011). Hence it is reasonable

to conduct research up on such area. Current study examines the impact of firm level

characteristics (size, leverage, tangibility, risk, growth, liquidity and age) on performance of

listed life insurance companies of Pakistan over seven years from 2001 to 2007.

The results of Ordinary Least Square (OLS) regression analysis indicates that size, risk and

leverage are important determinants of performance of life insurance companies of Pakistan

while ROA has statistically insignificant relationship with growth, profitability, age and

liquidity.

11

The review of related literature is divided in to four sections; the first section deals with the

concept of insurance and their profitability, the second section provides studies concerning

profitability performance of insurance companies together with other financial institutions.

The third section presents previous investigations on determinants of profitability in

insurance companies. The last section summarizes empirical literature concerning factors

affecting profitability in insurance companies.

2.2 The concept of insurance companies and their financial performance

Renbao Chen et.al (2004) stated in their investigation that “higher profits provide both the

means (greater availability of finance from retained profits or from the capital market) and

the incentive (a high rate of return) for new investment”. Therefore, we can understand from

the above explanation that insurance companies have double responsibility: in one way they

are required to be profitable so as to have high rate of return for new investment. On the

other hand, insurance companies need to be profitable in order to be solvent enough so as to

make other industries in the economy as they were before even after risk occurred.

Profitability is one of the most important objectives of financial management because one

goal of financial management is to maximize the owner` s wealth and profitability which in

turn indicates better financial performance. According to Hifza Malik (2011) insurance plays

a crucial role in fostering commercial and infrastructural businesses. From the latter

perspective, it promotes financial and social stability; mobilizes and channels savings;

supports trade, commerce and entrepreneurial activity and improves the quality of the lives of

individuals and the overall wellbeing in a country. Michael Koller (2011) in his investigation

12

identified that insurance companies are playing the role of transferring risk channeling funds

from one unit to the other (financial intermediation) such as general insurance companies and

life insurance companies respectively. This implies that insurance companies are helping the

economy of a country one way by transferring and sharing of risk which can create

confidence over the occurrences of uncertain event and in another way insurance companies

like other financial institutions plays the role of financial intermediation so as to channel

financial resources from one to the other.

Therefore, we can divide insurance companies in to two broad categories based on their role

to the economy; the general insurance companies and life insurance companies. For instance,

Renbao Chen et.al (2004) summarized firm specific factors affecting property/liability which

is general insurers and life/health insurance profitability separately that again provide

valuable guidelines for insurers financial health. This is because life/health insurance

companies are different from property/liability insurers in terms of operation, investment

activities, vulnerability and duration of liabilities. Life insurers are said to function as

financial intermediaries while general insurers function as risk takers Renbao Chen et..al

(2004)

2.3 The concept of profitability

According to Hamdan Ahmed Ali Al-Shami (2008) there are different ways to measure

profitability such as: ROA, return on equity (ROE) and return on invested capital (ROIC).

ROA is an indicator of how profitable a company is relative to its total assets. It gives us an

idea as to how efficient management is in using its assets to generate earnings whereas ROE

measures a company‟s profitability which reveals how much profit a company generates with

13

the money shareholders have invested. ROIC is a measure used to asses a company‟s

efficiency in allocating the capital under its control in profitable investments. This measure

gives a sense of how well a company is in using its money to generate returns. Comparing a

company‟s ROIC with its weighted average cost of capital (WACC) reveals whether invested

capital is used efficiently or not. In contrast, William H. Greene and Dam Segal (2004)

argued that the performance of insurance companies in financial terms is normally expressed

in net premium earned, profitability from underwriting activities, annual turnover, return on

investment, return on equity. These measures could be classified as profit performance

measures and investment performance measures. However, most researchers in the field of

insurance and their profitability stated that the key indicator of a firm‟s profitability is ROA

defined as the before tax profits divided by total assets. Philip Hardwick and Mike Adams

(1999), Hafiz Malik (2011) are among others, who have suggested that although there are

different ways to measure profitability it is better to use ROA.

Therefore, being profitable means that insurance companies are earning more revenues than

being disbursed as expenses. As explained above just to analyze the drivers of profitability, it

is useful to decompose either the return on asset ROA or ROE into their main components.

According to Swiss Re (2008) Profits are determined first by underwriting performance

(losses and expenses, which are affected by product pricing, risk selection, claims

management, and marketing and administrative expenses); and second, by investment

performance, which is a function of asset allocation and asset management as well as asset

leverage. The first division of the decomposition shows that an insurer‟s ROE is determined

by earnings after taxes realized for each unit of net premiums (or profit margin) and by the

14

amount of capital funds used to finance and secure the risk exposure of each premium unit

(solvency). That is why most researchers use ROA as a measure of profitability in financial

institutions.

The term profit can take either its economic meaning or accounting concept which shows the

excess of income over expenditure viewed during a specified period of time. On one hand,

profit is one of the main reasons for the continued existence of every business organization.

On the other hand, profit is expected so as to meet the required return by owners and other

outsiders. John J. Hampton (2009) clarified profitability ratio as a class of financial metrics

that are used to assess a business‟s ability to generate earnings as compared to its expenses

and other relevant costs incurred during a specific period of time. Accordingly, the term

'profitability' is a relative measure where profit is expressed as a ratio, generally as a

percentage. Profitability depicts the relationship of the absolute amount of profit with various

other factors. Similarly, Michael Koller (2011) argued that profitability is the most important

and reliable indicator as it gives a broad indicator of the ability of an insurance company to

raise its income level. In practice, executives define profits as the difference between total

earnings from all earning assets and total expenditure on managing entire asset-liabilities

portfolio Kaur and Kapoor, (2007).

The variation of profit among insurance companies over the years in a given country would

result to suggest that internal factors or firm specific factors play a crucial role in influencing

their profitability. It is therefore imperative to identify what are these factors as it can help

insurance companies to take action on what will increase their profitability and investors to

15

forecast the profitability of insurance companies in Ethiopia. To do so, it is better to see what

factors were considered in previous times by different individuals. The following points are

some of the work of others among many others.

2.4 Determinates of profitability in insurance companies: an empirical review

Profitability in insurance companies could be affected by a number of determining factors.

These factors, as explained above could be further classified as internal, industry, and

macroeconomic factors. However, as will be discussed in the coming consecutive sections of

the review, in most literatures, profitability with regard to insurance companies usually

expressed in as a function of internal determinants. Rather, most researches concerning

determinants of profitability in insurance companies are divided in to two, such as

determinants of profitability in property/liability or general insurance companies and in

life/health insurance companies. Accordingly, Hifza Malik (2011) in pakistan, Sylwester

Kozak (2011) in poland, Hamadan Ahamed Ali Al-Shami (2008) in United Arab emirates

(UAE), Swiss Re (2008) in Egypt and Jay Angoff Roger Brown (2007) in United kingdom

conducted their research concerning determinants of profitability in general insurance

companies where as Naveed Ahmed, Zulfqar Ahmed, Ahmad Usman (2011), in Pakistan,

Adams M., Hardwick P. and Zou H., (2008) in Canada, Desheng Wu Z., Sandra V. & Lianga

(2007), Wright, K. M. (1992), and others conducted their study on determinants of life and

health insurance companies. Hence, most of the researchers focused on internal factors

affecting profitability and most of the factors considered are age of company, size of

company, leverage ratio, growth rate, volume of capital, tangibility of assets and liquidity

ratio. Now let us see empirical evidences for each variable independently.

16

2.4.1 Internal Determinants

The internal determinants of insurance companies‟ profitability are those management

controllable factors which account for the inter-firm differences in profitability, given the

external environment. Accordingly Hafiz Malik (2011) defines internal determinants of

profitability as factors that could be influenced by management decisions. As stated by

Hamadin Ahamed Ali-Al-Alshami (2011) internal determinants can be broadly classified into

two sub-categories namely financial statement variables and non-financial statements

variables. The financial statement variables are determining factors which are directly driven

from items in a balance sheet and profit & loss accounts of the insurance companies. On the

other hand, the non-financial statement variables are those factors which are not directly

displayed on the financial statements accounts.

According to Yuqi Li (2007) financial institutions‟ non-financial statements variables are

classified as management quality, efficiency and productivity, age and number of branches.

Most researches concerning insurance companies are conducted with respect to only

financial statement variables. Hence, Hamadin Ahamed Ali Al-Shami (2008) in his

dissertation regarding UAE used financial statement variables such as size, leverage,

liquidity, tangibility of assets, volume of capital, and premium growth. Similarly, Hafiz

Malik (2011) in Pakistan used such variables mentioned above and age as a non-financial

statement variable. Sylwester Kozak (2011) in poland, Hamadan Ahamed Ali Al-Shami

(2008) in United Arab emirates (UAE), Swiss Re (2008) in Egypt and Jay Angoff Roger

Brown (2007) in United kingdom, Naveed Ahmed, Zulfqar Ahmed, Ahmad Usman (2011),

in Pakistan, Adams M., Hardwick P. and Zou H., (2008) in Canada, Desheng Wu Z., Sandra

17

V. & Lianga (2007), Wright, K. M. (1992), Flamini et..all (2009) in Sub-Saharan countries

are among others used financial statement variables as independent variables. The following

are the variables used in researches concerning profitability of insurance companies and

related financial institutions and the details of internal financial statement and one non

financial statement variable are discussed in detail in this section.

2.4.1.1 Firm size and age

Newly established banks are not particularly profitable in their first years of operation, as

they place greater emphasis on increasing their market share, rather than on improving

profitability Athanasoglou et al., (2005). Similarly, Yuqi li (2007) indicate that older banks

expected to be more profitable due to their longer tradition and the fact that they could build

up a good reputation. Obviously, the above empirical studies those include age as one of their

explanatory determinant indicates a positive relationship between age and profitability.

Several studies have been conducted to examine the effect of size and age on firm

profitability. However, the empirical evidences of the linkage between profitability and firm

size are somewhat inconsistent. For example, evidence collected by Philip Hardwick and

Mike Adams (1999) from UK companies suggests that there is an inverse relation between

profitability and firm size. Jay Angoff Roger Brown (2007) found that there is a positive and

significant relationship between the age of a company and its profitability as measured by

ROA. Similarly, the research conducted on the relationship among firm characteristics

including size, age, location, industry group, profitability and growth by Swiss Re (2008)

indicated that larger firms are found to grow faster than smaller and younger firms found to

grow faster than older firms. In contrast, Hamadan Ahamed Ali Al-Shami (2008) found no

18

significant statistical relation between age and profitability of insurance companies in UAE

but there exist a positive and statistical significant relation between firm size and

profitability. Similarly, Hafiz Malik (2011) in his Pakistan study found that there is

significantly positive association between age & size of the company and profitability. The

older the firm the more may be the profitability of the firm. This could be justified as

experience and efficiency in the operation process may decrease cost of production and he

found even that age is the strongest determinant of profitability.

In most literatures the effect of size on banks profitability are represented by total asset.

Flamini et.al (2009) indicated that size is used to capture the fact that larger firms are better

placed than smaller firms in harnessing economies of scale in transactions and enjoy a higher

level of profits. One of the most important questions underlying bank policy is which size

optimizes bank profitability. According to Athanasoglou et al., (2005) the effect of a growing

size of a bank on profitability has been proved to be positive to a certain extent.

Consequently, a positive relationship is expected between size and profitability by many

insurance area researchers. However, for firms that become extremely large, the effect of size

could be negative due to bureaucratic and other reasons Yuqi Li (2007). Hence, the size-

profitability relationship may be expected to be non-linear. Therefore most studies use the

real assets in logarithm and their square in order to capture the possible non-linear

relationship. Athanasoglou et al. (2005 and Yuqi Li found positive relationship between size

and profitability.

19

2.4.1.2 Liquidity

Liquidity from the context of insurance companies is the probability of an insurer to pay

liabilities which include operating expenses and payments for losses/benefits under insurance

policies, when due then shows us that more current assets are held and idle if the ratio

becomes more which could be invested in profitable investments. For an insurer, cash flow

(mainly premium and investment income) and liquidation of assets are the main sources of

liquidity Renbao Chen and Kie Ann Wong (2004). Empirical evidences with regard to

liquidity revealed almost inconsistent results. For instance, Naveed Ahmed et.al. (2011) in

his investigation in Pakistan found that ROA has statistically insignificant relationship with

liquidity. Similarly, several other studies also have been conducted to measure the

performance of the insurance companies. In contrast, Chen and Wong (2004) examined that,

liquidity is the important determinants of financial health of insurance companies with a

negative relationship. Similarly, Hakim and Neaime (2005) observed that liquidity, current

capital and investment are the important determinants of banks profitability. Valentina

Flamini, Calvin McDonald, and Liliana Schumacher (2009) in their investigation regarding

Sub-Saharan countries found significant and negative relationship between bank profitability

and liquidity.

2.4.1.3 Leverage

The trade of theory suggests a positive relationship between profitability and leverage ratio

and justified by taxes, agency costs and bankruptcy costs push more profitable firms towards

higher leverage. Hence more profitable firms should prefer debt financing to get benefit from

tax shield. In contrast to this pecking order theory of capital structure is designed to

minimize the inefficiencies in the firms’ investment decisions. Due to asymmetric

20

information cost, firms prefer internal finance to external finance and, when outside

financing is necessary, firms prefer debt to equity because of the lower information

costs. The pecking order theory states that there is no optimal capital structure since

debt ratio occurs as a result of cumulative external financing requirements. Insurance

leverage could be defined as reserves to surplus or debt to equity. The risk of an insurer may

increase when it increases its leverage. Literatures in capital structure confirm that a firm‟s

value will increase up to optimum point as leverage increases and then declines if leverage is

further increased beyond that optimum level.

For instance Renbao Chen and Rie Ann Wong (2004) stated that leverage beyond the

optimum level could result in higher risk and low value of the firm. Empirical evidences with

regard to leverage found to be statistically significant relationship but negative. For instance

Renbao Chen and Kie Ann Wong (2004), in Canada, Hamadan Ahamed Ali Al-Shami (2008)

in UAE, Hifza Malik (2011) in Pakistan, Sylwester Kozak (2011) in UK Swiss Re (2008) in

Egypt and Flamini et..al (2009) in Sub-Saharan countries found that negative but statistically

significant relationship between leverage and profitability of firms. Harrington (2005) stated

that the relationship between leverage and profitability has been studied extensively to

support the theories of capital structure and argued also that insurance companies with lower

leverage will generally report higher ROA, but lower ROE. Since an analysis for ROE pays

no attention to the risk associated with high leverage.

2.4.1.4 Volume of Capital

In most of the studies concerning insurance companies‟ volume of capital measures as the

difference between total assets and total liabilities and in some cases it is measured by the

ratio of equity capital to total asset. Insurance companies‟ equity capital can be seen in two

21

ways. Narrowly, as stated by Uhomoibhi T. Aburime (2008), it can be seen as the amount

contributed by the owners of an insurance (paid-up share capital) that gives them the right to

enjoy all the future earnings. More comprehensively, it can be seen as the amount of owners‟

funds available to support a business. The later definition includes reserves, and is also

termed as total shareholders‟ funds. No matter the definition adopted, volume of capital is

widely used as one of the determinants of insurance companies‟ profitability since it

indicates the financial strength of the firm. As it has been expected positive relationship

between profitability and capital has been demonstrated by Athanasoglou et al. (2005).

Studies conducted in different countries found that for non-life insurance companies, size of

capital is one of the important factors that affect ROA, Hifza Malik (2011) examined the

relationship between volume capital and return on asset for Pakistan insurance industry and

found positive and statistically significant relationship between insurance capital and

profitability. Similarly Hamadan Ahamed Ali Al-Shami (2008), found in his investigation

that there exists a positive and significant relationship between volume of capital and

profitability of the UAE insurance companies.

2.4.1.5 Tangibility of assets

Tangibility of assets in insurance companies in most studies is measured by the ratio of fixed

assets to total assets. A recent study by Naveed Ahmed et.al... (2011) investigates the impact

of firm level characteristics on performance of the life insurance sector of Pakistan over the

period of seven years. For this purpose, size, profitability, age, risk, growth and tangibility

are selected as explanatory variables while ROA is taken as dependent variable. The results

of OLS regression analysis revealed that leverage, size and risk are most important

22

determinant of performance of life insurance sector whereas ROA has statistically more of

insignificant relationship with, tangibility of assets. However, Hafiz Malik (2011) found that

there exists a positive and significant relationship between tangibility of assets and

profitability of insurance companies and argued that the highest the level of fixed assets

formation, the older and larger the insurance company is. In contrast to this, Yuqi Li (2007)

in UK found no significant relationship between tangibility of assets and profitability of

insurance companies.

2.4.1.6 Growth Rate

Growth as measured by the percentage change in total assets or sometimes percentage

change in premiums of insurance companies is expected to positively related with

profitability of insurance companies in Ethiopia. Insurance companies having more and more

assets over the years have also better chance of being profitable for the reason that they do

have internal capacity though it depends on their ability to exploit external opportunities.

Emperical evidence by Naveed Ahmed et al (2011) in Pakistan, Yuqi Li (2007) in UK and

Hamadin Ahmed Ali Al-Shami (2008) in UAE of their investigation found a positive and

statistically significant relationship between growth and profitability of insurance companies.

2.5 Summary of the literature review

Most literatures focus on factors affecting profitability of banks rather than insurance

companies. Therefore, there are fewer literatures concerning insurance companies as

compared to banks. The existing literatures concerning insurance companies could be

classified into two: determinants of financial performance of life and non-life insurance

companies. Empirical evidences regarding determinants of insurance companies focused only

23

on internal factors such as age, size, leverage, growth, volume of capital, tangibility of assets

and liquidity. The results found by the researchers mentioned above in the empirical revealed

inconsistencies according to the country in which the research is conducted regarding some

variables.

24

Chapter Three: Research Design and Methodology

3.1. Introduction

This study examines the previous findings in the literature, though not in Ethiopia and

applies the results in current practical settings of the insurance companies in the country.

Therefore, this chapter provides the detail steps and procedures used to conduct the analysis

of factors affecting profitability of insurance companies in Ethiopia. It includes the approach

adopted to examine the effect of main determinants on profitability, the type of data used and

the techniques employed to collect the data, the sampling mechanism including sample size,

the methods utilized to manage and analyze the data, and the process of constructing

empirical model with identification and measurement of its components, measurement and

selection of variables, expected relations between the dependent and independent variables.

Accordingly, a deductive approach is adopted by constructing an empirical model and

hypothesizing its linear relationship between determinants and its dependent variable. The

methodology of carrying out this research is based on the objectives of the paper and the

availability of relevant information.

To comply with the objective of this research, the paper is primarily based on survey of

quantitative research, which constructed an econometric model to identify and measure the

determinants of insurance companies‟ profitability. Classical linear regression analysis based

on the results of multiple regression analysis is adopted to measure the effect of determinants

on insurance companies‟ profitability by using E-views software packages and statistical

software package for social sciences (SPSS).

25

3.2 Research Approach

In terms of investigative study there are two common approaches to business and social

research: one is deductive approach that develops theories and hypotheses followed by a

research strategy to test the hypotheses; and second inductive approach that finds data and

develops theories as a result of the data analysis Saunders et al, (2003) as cited by Yuqi Li

(2007). The deductive approach introduces a high level of objectiveness in research through

external observation insofar as the choice of questions and subsequent phrasings are not

subjective. In contrast, the inductive approach provides a high level of subjectiveness and a

number of theoretical possibilities based on the context of the individual research situation

Yuqi Li (2007).

This study examines the previous findings in the literature, and apply the model in Ethiopian

insurance companies. Therefore, a deductive approach is adopted by constructing an

empirical model and hypothesizing its collinear relationship between determinants and its

dependent variable: profitability of insurance companies in Ethiopia.

3.3 Research Method

The methodology of carrying out this research is based on the objectives of the paper and the

availability of relevant information. To comply with the objective of this research, the paper

is primarily based on quantitative research, which constructed an econometric model to

identify and measure the determinants of profitability. Specifically, multiple regression

analysis is adopted to measure the effect of determinants on profitability. The use of multiple

regressions considers the simultaneous relationships amongst the multiple numbers of

independent and dependant variables found across the regression model, therefore suited to

the nature of the study.

26

The significance of the impact of the independent variables on dependent variables is, at the

same time, highlighted in using multiple regressions. Multiple regressions are further utilized

to examine the associative relationships between variables in terms of the relative importance

of the independent variables and predicted values of the dependent variables.

For the initial construction of the decomposed model an exploratory study was carried out

through a search of the available literature to identify the exact components of the model.

Further literature search was conducted to find other factors which could potentially and

clearly affect profitability of insurance companies in Ethiopia. By summarizing previous

studies, liquidity, volume of capital, firm size, age, leverage, growth and tangibility of assets

are selected to be included as explanatory variables in the model.

3.4 Data and Data Sources

To comply with the research objectives, the researcher focused on secondary data, which are

obtained from annual reports of individual insurance companies and NBE. And this is

because the advantage of using secondary data includes the higher quality data compared

with primary data collected by researchers themselves Stewart and Kamins, (1993) as cited

by Yuqi Li (2007); the feasibility to conduct panel evidence, which is the case in this study;

and the permanence of data, which means secondary data generally provide a source of data

that is both permanent and available in a form that may be checked relatively easily by

others, i.e more open to public scrutiny. Therefore, enhance the reliability of the data.

The principal secondary data sources for this paper are individual insurance companies

annual reports that contain detailed consolidated balance sheets and income statements and

27

National Bank of Ethiopia, which can provide comprehensive database for all insurance

companies.

The data collected and analyzed is a balanced panel of nine insurance companies in Ethiopia

operating over the last 9 years. Panel data is selected by the researcher in order to meet the

research objectives as it best fits better than the single time series or cross-sectional alone.

That is why Chris Brookes (2008) in his book clearly presents the advantage of using panel

data in the following way.

First, and perhaps most importantly, we can address a broader range of issues and tackle

more complex problems with panel data than would be possible with pure time-series or pure

cross-sectional data alone. Second, it is often of interest to examine how variables, or the

relationships between them, change dynamically (over time). To do this using pure time-

series data would often require a long run of data simply to get a sufficient number of

observations to be able to conduct any meaningful hypothesis tests. But by combining cross-

sectional and time series data, one can increase the number of degrees of freedom, and thus

the power of the test, by employing information on the dynamic behavior of a large number

of entities at the same time. The additional variation introduced by combining the data in this

way can also help to mitigate problems of multi-collinearity that may arise if time series are

modeled individually. Third, structuring the model in an appropriate way, we can remove the

impact of certain forms of omitted variables bias in regression results.

Panel data analysis is an increasingly popular form of longitudinal data analysis among social

and behavioral science researchers Yuqi Li (2007). Panel data analysis is a method of

studying a particular subject within multiple sites, periodically observed over a defined time

28

frame. With repeated observations of enough cross-sections, panel analysis permits the

researcher to study the dynamics of change with short time series.

Therefore, the combination of time series with cross-sections can enhance the quality and

quantity of data in ways that would be impossible using only one of these two dimensions.

3.5 Sampling Mechanism

Given the research objectives coupled with research questions, it is considered that purposive

sampling is employed so as to include all insurance companies established and serving with

in the specified period of time from June 2003 to June 2011 and the size for sample is nine

insurance companies operating over the period of 9 years as taken form table 1 above and the

rest of insurance companies will not have a chance to be included. Nine years is assumed to

be relevant because five years and above is the recommended length of data to use in most

finance literatures.

3.6 Data Analysis

Data analysis section of this study is based on descriptive analysis and regression analysis. It

means that the this section provides the descriptive analysis of the panel data and variables

for the study in collaboration with some important test such as normality of data, discusses

the correlation analysis between dependent and independent variables, deals the results of the

linear regression and data analysis that constitute the main findings of this study.

3.6.1. Descriptive Analysis

The descriptive statistics explores and presents an overview of all variables used in the

analysis. In this section the mean, minimum, maximum, standard deviation of the variables

are produced for the variables under study for the period 2003 to 2011.

29

3.6.2. The Correlation Analysis

This section shows how variables are related with each other. The results of this analysis

represent the nature, direction and significant of the correlation of the variables considered

under this study.

3.6.3. Regression Analysis

The regression analysis is used to examine the relationship between the profitability of

Ethiopian insurance companies and explanatory variables such as age, size, leverage,

liquidity, volume of capital, growth and previous year profitability of the companies

The result of a regression analysis is an equation that represents the best prediction of a

dependent variable from several other independent variables. In terms of regression analysis,

as panel data is adopted in this study, corresponded regression model is selected from fixed

effect and random effect regression. Fixed effects regression is the model to use when

researcher want to control for omitted variables that differ between cases but are constant

over time. It allows using the changes in the variables over time to estimate the effects of the

independent variables on dependent variable. Otherwise random effect estimation model is

used and it is the models to use when researchers want to control for omitted variables that

change over time but are constant between cases. It allows using the variation between cases

to estimate the effect of the omitted independent variables on dependent variable.

3.7 Design of Empirical Model

The literature generally, in so far as it is discussed, comes to the conclusion that the

appropriate functional form for testing is a linear function although there are dissenting

30

opinions. The Davidson, Godfrey, MacKinnon (1985) as cited by Swiss Re (2008)

specification test was also applied with results that supported the use of the linear function.

The regression model is used to identify the relationship between the profitability of

insurance companies and age of company, leverage ratio, growth, company size and volume

of capital. Data analyzed are with one dependent variable (profitability) and seven

independent variables (age of companies, size of companies, leverage ratio, volume of

capital, tangibility of assets, liquidity and previous profitability).

The generally accepted way of choosing between fixed and random effects is running a

Hausman test. Random effects will give better P-values as they are a more efficient

estimator, so random effects regression should be adopted if it is statistically justifiable to do

so. Based on Hausman test result, as shown in Appedix 2, the model is estimated through

random effect regression. The Hausman test checks a more efficient model against a less

efficient but consistent model to make sure that the more efficient model also gives

consistent results. It tests the null hypothesis that the coefficients estimated by the efficient

random effects estimator are the same as the ones estimated by the consistent fixed effect

estimator. If they are (insignificant P-value, Prob. of Chi-Square larger than .05) then it is

safe to use random effects and vice visa

For estimation purposes, the following general linear model is used:

ROAi,t =α+ ΣβjXji,t+υi----------------------------------------------------------------------------(1)

Where ROAi,t is the return on assets of insurance i for period t; α is the regression

constant;Xji,t denote insurance specific determinants; νi,t= εi,t is the disturbance term.

31

By using the model and comparing the co efficiency of each explanatory variable, it will

generate the finding that which factor is more significant in relation to insurance companies‟

profitability and the finding will correspond to the evidence in the literature.

3.8 Variable Selection and Measurement

As previously mentioned, the empirical part of this paper attempts to examine the main

determinants of profits of insurance companies‟ measurement of profitability. According to

Hamadan Ahamed Ali Al-Shami (2008), three important measures of firm‟s performance are:

profitability, size and survivorship. Profitability indicates the firm‟s ability to achievement of

the rate of return on a company‟s assets and investment funds. With regard to size, it is

revealed in his work as “a firm‟s ability to expand its size could be a reflection of it success

as earnings are reinvested and external funding could be easily found”. Whereas survivorship

indicates the ability to earn sustainable development concerning competitive advantages

beyond initial opportunities like an economic upturn or the early growth stage of an industry.

This research is concerning only on profitability of insurance companies in Ethiopia as a

financial performance and the internal factors that determine profitability. Hence, seven

characteristics are used as internal determinants of performance. Referring to previous

studies, the use of ratio in measuring leverage, liquidity, tangibility and profitability

performance is common in the literature of finance and accounting practices. Hafiz Malik

(2011) Hamadin Ali AL-Shami (2008) and Chen and Naveed Ahmed (2011) used ratio in

measuring insurance companies financial performance. The greatest advantage for using ratio

index in measuring performance is that it compensates disparities created by size Yuqi Li

(2007).

32

In line with earlier studies that examined the determinants of insurance companies‟

profitability, accounting ratios are used as measurement of individual variables. In specific,

the dependent variable, profitability of insurance companies, is measured by ROA. In order

to select the determinants as explanatory variables in the model, previous studies have also

been reviewed and literature suggests that the following factors exert strong impact on

insurance companies‟ profitability as internal determinants; therefore, they are adopted in the

constructed model. And following is the details of variables selected.

Profitability

There are many different ways to measure profitability, as shown in previous studies. In this

study net income before tax to total assets (ROA) is used to measure profitability, because

most of the studies regarding the subject used this ratio to determine the profitability of

insurance companies.

Age of company

This variable is measured by the number of years from the date of establishment until 2003,

2004, 2005, 2006, 2007, 2008, 2009, 2010 and 2011 for nine consecutive years.

Volume of capital

Previous studies used the book value of equity as a measure of volume of capital. Similarly

book value of equity is taken as a measure of volume capital for this study. Total equity

capital= book value of equity measured by the natural logarithm of book value of equity.

Company size

In different studies, different researchers use different measurements of company size such as

number of employees and total assets of a company. However, most of the researchers use

33

the log value of total assets as a measure of size in such area. Therefore, company size is

measured by total assets in log value.

Leverage

The amount of debt used to finance a company‟s assets. A company with significantly more

debt than equity is considered to be highly leveraged. This variable is measured by total debt

to total equity value of the company.

Firm growth

Growth is simply the change in size of the company as measured by the percentage change in

total assets.

Liquidity

Liquidity from the context of insurance companies is the probability of an insurer to pay

liabilities which include operating expenses and payments for losses/benefits under insurance

policies, when due and therefore, measured by total current assets to total current liabilities.

To capture the tendency of profits to be persistent over time (due to market structure

imperfections or high sensitivity to auto-correlated financial factors), the researcher tried to

adopt a dynamic specification of the model, with a lagged dependent variable among the

regressors. Cheris Brooks (2008) in his book for introductory econometrics for finance

argued that lagged values of variables may capture important dynamic structure in the

dependent variable that might be caused by a number of factors such as inertia of the

dependent variable and overreactions. This yields the following model specification:

ROAi,t = α+ γROAi,t-1 + Σβj,t+υi----------------------------------------------------------------(2)

34

Where ROAi,t-1 is the one period lagged profitability and γ measures the speed of mean

reversion. A value of delta between 0 and 1 indicates that profits are persistent, but they will

eventually return to the equilibrium level. Specifically, values close to zero denote a high

speed of adjustment and imply relatively competitive market structure, while a value closer

to 1 implies slower mean reversion, and therefore, less competitive markets.

Taking all these explanatory variables into consideration, the extended equation to reflect the

variables is formulated as follows:

ROAi,t= α + γROAi,t-1+β0Agei,t + Lnβ1 Sizei,t + β2 Levi,t + β3Gri,t + Lnβ4Voci,t +

5TAi,t+6LQi,t+εi,t-----------------------------------------------------------------------(3)

Where:

1. ROAi,t is the profitability in insurance company i at time t (dependent variable) in this

study return on assets (The return on assets (ROA) defined as the insurance companies‟

before tax profit over total assets) is used to measure profitability. My justification is that

ROA as the key proxy for insurance companies‟ profitability, instead of the alternative

return on equity (ROE), because an analysis of ROE disregards financial leverage and the

risks associated with it as a measure of profitability in insurance companies. Since profits

are a flow variable generated over the years, as opposed to the stock of total assets, I

measure this ratio as a running year average, with the average value of assets of

consecutive years as a denominator.

2. α is constant,

3. ROAi,t-1: the profitability of insurance company „i‟ in the previous times „t‟

35

4. Age: the variable age of company will be measured from the number of years to date of

establishments (difference between observation year and establishment year) or in other

words the age of each insurance company at time „t‟

5. Size: company size will be measured by total assets in log value,

6. Lev: is leverage ratio and for this variable the proxy is the ratio of total debt to equity

value of the company that means total debts divided by total equity

7. Voc: is volume of capital and it is measured as the book value of equity so will also use

the book value of equity as a measure of capital (total equity capital that is book value of

equity will be measured by the natural log of book value of equity) and

8. TA: Tangibility (Fixed assets divided by total assets)

9. LQ: Liquidity (Current assets divided by current liabilities)

10. 0…6: coefficient of independent variables

11. ε is error term.

12. i is insurance companies 1 to 9

Based on review of relevant and related literatures, it is hypothesized that volume of capital,

growth, age and size of company, leverage ratio, liquidity ratio and previous profitability are

expected to influence firms‟ profitability as measured by ROA. Accordingly, the following

hypotheses are tested by the study:

H1: There is a positive relationship between age and profitability of insurance

companies in Ethiopia.

H2: There is a positive relationship between size and profitability of insurance

companies in Ethiopia.

36

H3: There exists a positive relationship between any increase in volume of capital and

profitability of insurance companies in Ethiopia.

H4: There is a negative relationship between leverage and profitability for Ethiopian

insurance companies.

H5: There is a positive relationship between growth and profitability of Insurance

companies in Ethiopia.

H6: Tangibility of assets of insurance companies and their profitability are negatively

related.

H7: Liquidity ratio and profitability of insurance companies are negatively related.

Accordingly, in order to test for the empirical relevance of the hypotheses regarding to the

determinants of insurance profitability, based on other studies the model with respect to

determinants of profitability in insurance companies is designed as in the following diagram.

Based on the hypotheses above, the following table shows the expected results of each

independent variable

To summarize, this chapter deals the approach adopted to examine the effect of main

determinants on profitability, the type of data used and the techniques employed to collect

the data, the sampling mechanism including sample size, the methods utilized to manage and

analyze the data, and the process of constructing empirical model with identification and

measurement of its components, measurement and selection of variables, expected relations

between the dependent and independent variables.

37

Chapter 4: Analysis and Findings

4.1 Introduction

This chapter presents the empirical test results based on the linear regression to test the

outcomes of the analysis for nine insurance companies in Ethiopia during the period of 2003

to 2011. The investigation is with regard to the relationship between profitability as

dependent variable and age of insurance companies, size of insurance companies, volume of

capital, leverage ratio, growth rate, tangibility of company assets and liquidity ratio as

independent variables. Therefore, this chapter provides the results from the analysis of data

and its interpretation. This chapter is divided into four sections. The first section provides the

analysis throughout, test of the normality of data; the second section presents descriptive

analysis of the data and variables for the study; the third section discusses the correlation

analysis between dependent and independent variables followed by testing the hypothesis in

the fourth section; the fifth section lays down the results of regression analysis that constitute

the main findings of this study and presents the application of the model and eventually the

summary of the chapter is presented in the last section.

Normality of Data

According to Gujarati (1995) before running regression analysis, it should be noted that there

are four classic assumptions in undertaking the regression analysis and one of them is

normality of data. Therefore, normality test becomes relevant. Chris Brooks (2008) also

noted that in order to conduct hypothesis test about the model parameter, the normality

assumption must be fulfilled. The normality assumption is about the mean of the residuals is

zero. Therefore, the researcher used graphical methods of testing the normality of data as

shown below.

38

From figure 4.1 below, it can be noted that the distribution is normal curve, indicating that

the data confirms to the normality assumption. In addition, the normal probability plots were

used to test the normality of data as shown below in figure 4.1 and figure 4.2.

Figure 4.1: Histogram

Source: SPSS regression output

If the residuals are normally distributed around its mean of zero the histogram is a bell-

shaped. The shape of the histogram as shown above in figure 4.1 revealed that the residuals

are normally distributed around its mean of zero.

39

Figure 4.2: Normal P-P plot of regression standardized residual

Source: SPSS regression output

Similarly, the above figure shows the normal distribution of residuals around its mean of

zero. Hence the normality assumption is fulfilled as required based on the above two figures,

it is possible to conclude that the inferences that the researcher will made about the

population parameter from the sample is somewhat valid.

Descriptive statistics

Univariate analysis of all the variables in this study is represented as in the following table.

In this section, the study presents the empirical test results that include the descriptive. It

explores and presents an overview of all variables used in the study.

The table below (Table 4.1) shows that there are 81 numbers of valid cases or “N” for each

variable.

40

Table 4.1 Descriptive statistics

Mean Std. Deviation N

ROA 0.060254 0.0425792 81

LAG 0.062355 0.0445553 81

AGE 13.67 7.440 81

SIZE 18.720871 1.0561071 81

LEV 2.1274820 0.9355457 81

GR 0.181054 0.1809098 81

VOC 17.62280 0.8792137 81

TA 0.134005 0.0880514 81

LQ 1.968611 0.6913957 81

Source: SPSS descriptive statistics out put

The table indicates that the mean values of all the variables ranges from minimum of 0.06 for

ROA to a maximum of 18.7 for size. The average profitability as measured by ROA for

Ethiopian insurance companies during the study period is about 0.06 and the value of the

standard deviation for ROA is 0.04 which implies the presence of moderate variations among

the values of profitability across the insurance companies included for this study.

The mean value of age is 13.67 years and there are significant differences among values of

age because the value of the standard deviation as shown in the table is 7.44 years.

The mean value of size is 18.72. Therefore, with regard to size as shown in the table above,

there exists significant variation across the sample insurance companies for the reason that

the value of the standard deviation is 1.056107. Hence the highly variated size among

41

insurance companies my have significant impact on profitability of insurance companies that

we are going to se in the regression results

The mean value of leverage is 2.127482 implies that there were moderate differences among

the values of leverage as measured by debt to equity ratio across the sample insurance

companies under this study and is because the value of standard deviation is 0.935546.

From table 4.1 above, the mean value of growth is 0.181054 and the value of standard

deviation for the same variable is 0.180910 which shows that there were no significant

variations among the values of growth as measured by the change in total assets over the

years across the sample insurance companies.

The average value for volume of capital (VOC) has become 17.62280 with a standard

deviation of 0.879214. Therefore, there exists very significance variation among the values

of volume of capital across the sample insurance company included in this study.

Table 4.1 also shows that the mean value for tangibility of assets is 0.134005 and the

standard deviation is 0.088051 implies that there exists moderate variation among the values

of tangibility of assets in insurance companies.

Similarly the mean value of liquidity ratio is 1.968611 with the value of standard deviation

0.691396 which also shows us the existence of moderate difference among the values of

liquidity ratio for insurance companies under consideration. Therefore, this study is

conducted to what extent; the variations in factors affect the profitability of insurance

companies in Ethiopia. As indicated in appendix 3; profitability measured by ROA for

different insurance companies considered for this study for nine consecutive years is

different. Identification of the internal factors that affect the profitability of these companies

is the task of the researcher for this study.

42

Test for Hetroskedasticity

The other important assumption for classical linear regression model is that the disturbances

appearing in the population regression are homoscedastic that means the variance of the error

term is consistent. If errors don not have a constant variance (not homoscedastic), they are

said to be Hetroskedastic Chris Brooks (2008).

For the test of the presence of hetroskedasticity, the researcher used white test and is based

on the following null hypothesis and its respective alternative.

H0: There is no hetroskedasicity

H1: There is hetroskedasticity

To test the presence of hetroskedasticity, the residual sum of squares for each observation

have been calculated and regressed against the independent variables. The results obtained

are as follows:

Table 4.2 White test regression

Model Summary

Model R R Square Adjusted R Square Std. Error of the Estimate

1 0.382a 0.146 0.064 0.00138718154

Source: SPSS regression output

Table 4.3 Chi Square calculated and tabulated

Test t-statistics Χ2 calculated

=nR2

X2 (5% sig. level)

X2α(p), where p=xi+1

White‟s test 11.826 18.307

Source: SPSS regression output

43

The results from the table 4.2 and 4.3 above show that X2 square value obtained through

calculation is less than the value of Chi square value from the table at 5% significant level.

According to white test if the value of chi square calculated is greater than the chi square

tabulated at a given significant level, we have to reject the Ho of no heroskedasticity

otherwise we fail to reject it and accept the alternative that is there exists hetroskedasticity.

The t-statistics value (chi square calculated) from table 4.3 above is 11.826 which is less than

chi square tabulated at 5% significant level, 18.307. Hence 11.826 is less than 18.307 so that

we fail to reject the null hypothesis of no hetroskedasticity. In this case it is indicated that

there is no evidence for the existence of hetroskedasticity.

Test for autocorrelation

The Durbin-Watson Statistics (D-W stat.) from the regression result before analysis shows

that 1.89 which is approaching to 2 and hence no evidence for the presence of

autocorrelation. Of course autocorrelation test will be invalid with the presence of lagged

variable so tested with out the lagged variable and result for DW stat. becomes 1.75. In

addition the following distribution figure for residuals shows no pattern in residuals implies

that there is no autocorrelation.

44

Figure 4.3 Residuals distribution

The next step is an analysis of bivariate relationship between variables. This is shown by

using correlation matrix as derived from the E-views and SPSS. And it is because SPSS can

read files saved in the form of E-views.

Correlation Analysis

The correlation coefficient represents the linear relationship between two variables. The most

widely-used type of correlation coefficient is Pearson r, also called linear or product-moment

correlation. The significance level calculated for each correlation is a primary source of

information about the reliability of the correlation. The significance of a correlation

coefficient of a particular magnitude will change depending on the size of the sample from

which it was computed. Here, the analysis is with regard to significant correlations between

the dependent variable and each independent variable separately, to decide whether to accept

or reject the hypotheses.

Table 4.4 shows us correlations between ROA and independent variables. Return on assets is

negatively correlated with leverage (lev) and liquidity (LQ). The coefficient estimates of

correlation -0.015376 and -0.045734 for leverage and liquidity respectively. The result

suggests that leverage and liquidity are independent of return on assets.

-.12

-.08

-.04

.00

.04

.08

10 20 30 40 50 60 70 80

ROA Residuals

45

Table 4.4 Correlations between profitability and independent variables

ROA LAG AGE SIZE LEV GR VOC TA LQ

ROA 1.000000 0.259071 0.222132 0.356744 -0.044616 0.257726 0.436393 0.090525 -0.056834

LAG 0.259071 1.000000 0.155454 0.254125 -0.038359 -0.285424 0.316130 -0.032067 0.099973

AGE 0.222132 0.155454 1.000000 0.478159 0.575873 -0.195013 0.743681 -0.092996 0.060831

SIZE 0.356744 0.254125 0.478159 1.000000 0.672534 -0.054147 0.971563 -0.182436 -0.083522

LEV -0.044616 -0.038359 0.575873 0.672534 1.000000 0.076373 0.484409 -0.033366 -0.312444

GR 0.257726 -0.285424 -0.195013 -0.054147 0.076373 1.000000 -0.093431 -0.027489 0.020878

VOC 0.436393 0.316130 0.743681 0.971563 0.484409 -0.093431 1.000000 -0.217501 0.023037

TA 0.090525 -0.032067 -0.092996 -0.182436 -0.033366 -0.027489 -0.217501 1.000000 -0.435031

LQ -0.056834 0.099973 0.060831 -0.083522 -0.312444 0.020878 0.023037 -0.435031 1.000000

Source: SPSS Output

The significance level of this result is indicated in the appendix section. The highest positive

percentages are size as measured by total assets and volume of capital. The coefficients of

correlations are 37.67% and 43.64% respectively and they are positively correlated with

profitability as measured by ROA. This means that as these variables increase ROA also will

increase. The table also shows that age and tangibility are correlated positively but it is not

statistically significant at 1%, 5% and 10% significance level as shown in appendix 6.

Therefore, profitability is independent of age and tangibility of assets. The descriptive

statistics also indicate two of the variables namely size and volume of capital are strongly

correlated with each other with a coefficient estimate of 97%. Hence, there may appear

multicollinearity problem and care should be taken with the results of regression analysis.

46

The correlation analysis shows that ROA is significantly correlated with size of company,

leverage ratio, firm growth, volume of capital and liquidity ratio. The analysis also indicates

that several independent variables are correlated with each other. For instance volume of

capital is positively correlated with size of company and age of company as well. This

observation indicates that special attention should be given to the possible problem of

multicollinearity when regression analysis is executed.

Testing the hypotheses based on correlation analysis

Hypothesis 1

Table 4.5 correlation between age and ROA

Correlation coefficient (r) P-value

AGE 0.22

0.15

Source: SPSS output

Form the table above, we found that there is no significant relationship between age and

profitability as measured by ROA. Therefore, we do not accept the H1.

Hypothesis 2

Table 4.6 correlation between size and ROA

Correlation coefficient (r) P-value

Size 0.36

0.00

Source: SPSS output

From the table above we can see that size is positively correlated with ROA and this

relationship is statistically significant. Hence H2 is accepted.

47

Hypothesis 3

Table 4.7 correlation between leverage ratio and ROA

Correlation coefficient (r) P-value

Leverage ratio -0.044616

0.0025

Source: SPSS output

The results from table 4.5 above show that there is a significant and negative relationship

between leverage ratio and ROA and therefore H3 is accepted.

Hypothesis 4

Table 4.8 correlation between growth and ROA

Correlation coefficient (r) P-value

Firm Growth 0.257726

0.02002

Source: SPSS output

The results from table 4.6 show that there is a significant positive correlation between firm

growth and ROA. Hence H4 is accepted

Hypothesis 5

Table 4.9 Correlation between Volume of capital and ROA

Correlation coefficient (r) P-value

VOC 0.436393

0.0000

Source: SPSS output

The results from table 4.7 show that there is a significantly strong correlation between VOC

and ROA. Hence H5 is accepted.

48

Hypothesis 6

Table 4.10 Correlation between tangibility of assets and ROA

Correlation coefficient (r) P-value

Tangibility of assets 0.0905

0.4216

Source: SPSS output

The results from the table 4.8 show that there is no significant relationship between

tangibility of assets and ROA. Hence H6 is not accepted.

Hypothesis 7

Table 4.11 Correlation between liquidity ratio and ROA

Correlation coefficient (r) P-value

Liquidity -0.0597

0.0614

Source: SPSS output

The results of the table 4.9 show that there is slightly negative correlation between liquidity

and ROA. Hence H7 is accepted.

Collinearity statistics

In this section, the problem of multi-collinearity is discussed based on the results that have

been produced. Multi-collinearity is investigated using tolerance value and variance inflator

factor (VIF) value. An insignificant tolerance value indicates that the variable under

consideration is almost a perfect linear combination of the independent variables already in

the equation and that it should not be included to the regression equation. Tolerance ranges

from zero to one. The closer the tolerance value to zero indicates a level of multi-collinearity.

49

Table 4.12 Collinearity (model 1)

Model

Coefficients

t Sig.

Collinearity Statistics

B Std. Error Tolerance VIF

1 (Constant) -.375 .113 -3.309 .001

LAG .175 .089 1.969 .053 .777 1.287

AGE .001 .001 1.346 .182 .309 3.238

SIZE -.120 .070 -1.716 .090 .003 321.102

LEV .010 .020 .517 .606 .034 29.642

GR .110 .022 5.099 .000 .795 1.257

VOC .149 .071 2.100 .039 .002 445.438

TA .069 .047 1.472 .145 .719 1.391

LQ -.017 .007 -2.466 .016 .514 1.947

Source: Random effect regression output of SPSS

Regression Analysis

From table 4.12 for the first model the results show that VIF value is 445.438, 321.102 and

29.642 for size volume of capital and leverage respectively. It indicates that this model is not

free from multi-collinearity between the independent variables in this model. The correlation

analysis made here in this paper shows that volume of capital is highly correlated with size

and moderately correlated with leverage. Hence the model will be tested for the second time

by dropping out of the volume of capital form list of regressors.

50

Table 4.13 Collinearity (model 2)

Model 2 coefficients t Sig. Collinearity statistics

B Std.Error Tolerance VIF

C -0.395 .116 -3.415 .001

LAG .181 .091 1.999 .049 .778 1.285

AGE .001 .001 1.215 .228 .301 3.230

SIZE .026 .007 3.936 .000 .259 3.867

LEV -0.030 .006 -5.154 .000 .419 2.387

GR .104 .022 4.746 .000 .811 1.233

TA .067 .046 1.409 .163 .719 1.391

LQ -0.012 .077 -1.758 .083 .602 1.662

Dependent variable: ROA

Source: Random effect regression output of SPSS

From the table above, for the second model excluding volume of capital, from the list of all

regressors, the results show that VIF value for all variables becomes less and the tolerance

value for all variables is not near to zero. It indicates that this model is free from multi-

collinearity. Hence, there is no problem of multi-collinearity between the variables in this

model. Therefore regression analysis is done by excluding volume of capital from the model.

Table 4.14 Model summary (b)

Model R R2

Adjusted R2 Standard Error of the estimate

2 0.699a 0.488 0.439 0.0318967

Source: SPSS output

51

Hypothesis testing and interpretation of the results based on random effect panel

Shown below in table 4.3.1 is the empirical result of the study by using random estimators of

panel data using SPSS.

Table 4.15 panel random effect estimation result after excluding VOC

Dependent variable: ROA

Independent Variables

Panel random effect estimation result

β1 β2 Std.error Sig.

LAG 0.181 0.190 0.116 0.049***

AGE 0.001 0.183 0.001 0.228

SIZE 0.026 0.648 0.007 0.000*

LEV -0.030 -0.667 0.006 0.000*

GR 0.104 0.441 0.022 0.000*

TA 0.067 0.139 0.046 0.163

LQ -0.012 -0.190 0.077 0.083***

Observations = 81

R2

= 0.448

Adjusted R2 = 0.439

F-statistics = 9.937

DW statistics

81

0.48

0.439

9.937*

1.89

*Significant at 1%, ** Significant at 5%, *** Significant at 10%

Source: Random effect regression result of E-Views

The results of regression for six independent variables on ROA in model 2 are presented in

Table 4.14. This table shows the independent variables entered into the regression model.

When all the inter-correlation the variables are taken into account, the R square is 0.488, it

52

means that on average 48.8% of the variation in ROA can be explained by the independent

variables under the model above. However, t test shows that two of the independent variables

namely age and tangibility of assets are not significant with (P=0.1824) and (p=0.1455)

respectively. Hence this result is consistent with the correlation analysis

Table 4.16 ANOVA (b)

Model Sum of Squares df Mean Square F Sig.

2 Regression .071 7 .010 9.937 .000a

Residual .074 73 .001

Total .145 80

Source: Random effect regression result of E-Views

a. Predictors: (Constant), LQ, GR, SIZE, LAG, TA, LEV, AGE

b. Dependent Variable: ROA

The ANOVA table above shows that the F value is significant at p=0.000 when the seven

variables are entered together. The model explains the relationship between the independent

variables and the dependent variable, moreover this model is significant and use five

independent variables are predictors of the ROA.

The main purpose of observing the adjusted value of R square is to apprehend the best model

that can explain ROA in the Ethiopian insurance companies. It is noted from the regression

result that the adjusted R square in the second model is 0.439. This indicates the model is the

best to explain ROA of insurance companies in Ethiopia. Which means on average 43.9 % of

the change in profitability as measured by ROA can be explained by the variables in the

model. Hence the function for regression equation for second model is:

53

ROA=-0.3948+0.1813LAG+0.0261SIZE-0.0304LEV+0.1039GR-0.0117LQ+ɛ

Alternatively the model is also tested by excluding one of the highly correlated variables

(size) instead of VOC from the model as shown in the following tables

Table 4.17 Collinearity (model 3) random effect regression results excluding size

Model 3 Coefficients

t-stat.

Significance

Collinearity statistics

3 Variables β Std.

error

Tolerance

VIF

Constant -0.409 0.113 -3.614 0.001

LAG 0.175 0.090 1.946 0.055 0.777 1.287

AGE 0.001 0.001 1.150 0.254 0.313 3.199

LEV -0.023 0.005 -4.250 0.000 0.483 2.069

GR 0.104 0.022 4.830 0.000 .0815 1.228

TA 0.070 0.047 1.471 0.146 0.719 1.391

VOC 0.024 0.007 4.150 0,000 0.359 2.788

LQ -0.013 0.007 -1.919 0.059 0.610 1.640

Dependent variable: ROA

Source: SPSS Random effect regression output

Table 4.18 Model Summary

Model R R2

Adjusted R2

Std. Durbin-Watson

3 .706a .478 .420 .0315891 1.868

a. Predictors: (Constant), VOC, LQ, GR, LAG, TA, LEV, AGE

Source: Random effect results of E-views

54

From the above table we can conclude that volume of capital is one of the important

determining factors of profitability of insurance companies in Ethiopia. The regression

results show a regression coefficient of 0.024, t-statistics of 4.150 and p-value of 0.000.

Hence volume of capitals significantly and positively affects profitability of insurance

companies in Ethiopia and this result is consistent with the hypothesis of the study.

The result implies that a 1% increase in volume of capital will result in a 2.8% increase in

profitability. The coefficient of volume of capital is positive and highly significant, meaning

that well capitalized insurance companies experience higher returns of assets and hence

higher profits.

4.2 Summary of Findings

Discussion of findings is more depending on model two above and it is because the R square

for the second model is to some extent more than the third model. Hence model two explains

the study better than model three.

4.2.1 Age

In this study, random effect regression result shows that there is no significant relationship

between age of insurance companies and their profitability in Ethiopia. As shown above in

model two, the regression coefficient of age 0.001 with a t-statistics of 1.215 including

significance value of 0.228. Thus from the results we can conclude that there exists no

relationship between age and profitability of insurance companies in Ethiopia. Hence this

result is not consistent with the hypothesis of the study. A research previously conducted was

also resulted in inconsistent results some indicated that age is negatively related with

profitability. For instance Swiss Re (2008) in Egypt indicated that larger firms are found to

55

grow faster than smaller and younger firms found to grow faster than older firms. In contrast,

Hamadan Ahamed Ali Al-Shami (2008) found no significant relation between age and

profitability of insurance companies in UAE. Although the results show no statistical

significance between these variables, it can be concluded that the age of a firm still explains

the variation in profitability of insurance companies positively.

4.2.2 Size

The positive and significant coefficient of the size variable gives support to the economies of

scale market-power hypothesis. Larger insurance companies make efficiency gains that can

be captured as higher earnings due to the fact that they do not operate in very competitive

markets. The regression results by different researchers indicated that there exists a positive

relationship between size and profitability of firms. Swiss Re (2008) indicated that larger

firms are found to grow faster than smaller firms. In addition, Hamadan Ahamed Ali Al-

Shami (2008) found positive and statistical significant relation between firm size and

profitability. Similarly, Hafiz Malik (2011) in his Pakistan study found that there is

significantly positive association between size of the company and profitability.

Similar to most of the researchers mentioned above, in this study, the panel random effect

estimation result revealed that there exist a significant and positive relationship between size

and profitability of insurance companies in Ethiopia with a regression coefficients of 0.026,

t-statistics of 3.936 p-value of 0.000. Hence, the variables have statistically significant

positive relationship. Hence the result of the regression output is consistent with the

hypothesis of the study.

56

4.2.3 Leverage

The regression results of the study show that there is a statistically significant negative

relationship between leverage ratio of insurance companies and their profitability in Ethiopia

with a regression coefficient of -0.03, t-statistics of -5.154 and p-value of 0.000. Hence, the

results are consistent with the hypothesis of the study. Literatures in capital structure confirm

that a firm‟s value will increase up to optimum point as leverage increases and then declines

if leverage is further increased beyond that optimum level. Empirical evidences with regard

to leverage found to be statistically significant relationship but negative. For instance Renbao

Chen and Kie Ann Wong (2004) in Canada, Hamadan Ahamed Ali Al-Shami (2008) in UAE,

Hifza Malik (2011) in Pakistan, Sylwester Kozak (2011) in UK found that negative but

statistically significant relationship between leverage and profitability of firms.

4.2.4 Growth

Growth as measured by the percentage change in total assets is positively related with

profitability of insurance companies in Ethiopia. The results of the random effect regression

analysis show that there is a positive and statistically significant relationship between growth

rate and profitability of insurance companies with a regression coefficient of 0.104, t-

statistics of 4.746 and p-value of 0.000. Hence the results are consistent with the hypothesis

of the study and correlation analysis. Insurance companies having more and more assets over

the years have also better chance of being profitable for the reason that they do have internal

capacity though it depends on their ability to exploit external opportunities. Emperical

evidence by Naveed Ahmed et al (2011) in his investigation of Pakistan insurance companies

found a positive and statistically significant relationship between growth and profitability of

insurance companies.

57

4.2.5 Tangibility of assets

The regression results concerning tangibility of assets show that there is no statistically

significant relationship between tangibility of assets and profitability of insurance companies

in Ethiopia. The regression coefficient is 0.067, t-statistics 4.746 and p-value of 0.163. Hence

the result is inconsistent with the hypothesis of the study but consistent with the correlation

analysis. Although the statistical results reveal no significant relationship between the

variables, it can be concluded that tangibility of assets still positively explains profitability of

insurance companies in Ethiopia. Regarding the effect of tangibility of assets of companies

on their financial performance, empirical evidences by Hafiz Malik (2011) in Pakistan

revealed that there exists a positive and significant relationship between tangibility of assets

and profitability of insurance companies and argued that the highest the level of fixed assets

formation, the larger the insurance company is. Hence tangibility of assets is also part of the

size of the company.

4.2.6 Liquidity

For an insurer, cash flow (mainly premium and investment income) and liquidation of assets

are the main sources of liquidity Renbao Chen and Kie Ann Wong (2004) the larger the

liquidity ratio shows more current assets are held which would have been invested in

profitable business hence the more the liquidity ratio the lower is the profitability. Empirical

evidences by Chen and Wong (2004) in Canada examined that, liquidity is the important

determinants of financial health of insurance companies with a negative relationship.

Similarly, Hakim and Neaime (2005) observed that liquidity negativel related with

profitability. Valentina Flamini, Calvin McDonald, and Liliana Schumacher (2009) in Sub-

Saharan countries found significant but negative relationship between bank profitability and

58

liquidity. Consistent to the above empirical studies, the results of regression analysis show

that there exist a negative and somewhat statistically significant relation between liquidity

and profitability of insurance companies in Ethiopia. The results of the random effect

regression of panel evidence over nine insurance companies for nine years revealed

regression coefficient of -0.012, t-statistics of -1.758 and p-value of 0.083. Hence at 10%

significance level, liquidity ratio negatively explains profitability of insurance companies and

it is consistent with the hypothesis of the study.

4.2.7 Volume of Capital

From model three above as expected, it is indicated that positive relationship between

capital strength and profitability. The coefficient of the VOC is 0.024 and is relatively high

even at 1% significant level, showing that an increase in volume of capital will result in

increased profitability. This is in line with the expectation as an insurance company with a

sound capital position is able to pursue business opportunities more effectively and has more

time and flexibility to deal with problems arising from unexpected losses, thus achieving

increased profitability. Hence finding in this study is consistent with previous studies. For

instance, Hamadin Ahamed Ali-Alshami (2008) in UAE, Hafiz Malik (2011) in Pakistan

Yuqi Li (2007) in UK and indicates that well capitalized insurance companies face lower

costs of going bankrupt, which reduces their cost of funding or that they have lower needs for

external funding which results in higher profitability.

59

Regression coefficient of size at 0.0261 indicates that when firm size increases by 1%

the ROA will increase by 2.6%.

Regression coefficient of Lev at -0.0304 indicates that when leverage increases by

1% the ROA will decrease by 3.04%.

Regression coefficient of GR at 0.1039 indicates that when firm growth increases by

1% the ROA will increase by 10.39%

Regression results of Volume of capital indicates that as volume of capital increase

by 1% ROA will also increase by 2.4%

The regression coefficient of LQ at -0.012 indicates that when liquidity ratio

increases by 1% the ROA will decrease by 1.12%.

Regression coefficient of LAG at 0.1813 indicates that the one period lagged

profitability and 0.1813 measures the speed of mean reversion. This value lies

between 0 and 1 indicates that profits are persistent. The magnitude and significance

of the coefficient on the lagged ROA confirm the dynamic nature of the model, and

show a moderate persistence in return. The coefficient estimate of 0.18 suggests the

existence of market power in the Ethiopian insurance sector, but indicates profits tend

to adjust fairly fast to their average level. This result is consistent with those reported

in Athanasoglou, et al. (2005) Flamini et.al (2009) in Sub-Saharan countries. The lag

coefficient on the one-year lagged ROA is positive and highly significant, which

confirms the positive conditional serial correlation in returns that found in this model.

60

To summarize, this chapter presents the results of the hypotheses of the independent

variables tested on the dependent variable (ROA). The equation applied in this study is

examined against multi-collinearity.

Empirical results provide detailed discussions on sample descriptive statistics and mean

comparison between ROA and independent variables (age, size, leverage, growth, volume of

capital, tangibility of assets and liquidity ratio) followed by correlation analysis to determine

the relationship between dependent variable and towards independent variables. Regression

analysis is also used to describe the profitability among insurance companies.

ROA is affected positively by firm size, volume of capital and growth but negatively by

leverage and liquidity. Therefore, growth, leverage, size, volume of capital and liquidity are

identified as determinant factors of profitability in insurance companies of Ethiopia. The

findings of this study contribute towards a better understanding of financial performance in

Ethiopian insurance companies. ROA and seven other variables that represent age, size,

leverage, growth, volume of capital tangibility and liquidity were developed to test which

factors best describes profitability of Ethiopian insurance companies.

The results show that growth, leverage, volume of capital, size and liquidity are the most

important factors affecting profitability of insurance companies in Ethiopia respectively in

order of their degree of influence. The results show that there is no relationship between

profitability and age of company of the Ethiopian insurance companies. Similarly, the results

show that there exists no relationship between tangibility of assets and profitability of

insurance companies in Ethiopia.

61

Chapter Five: Conclusion and Implications of the Results

5.1 Conclusion

The objective of this study is to examine the internal factors affecting profitability of

insurance companies as measured by ROA. This study used secondary data during the period

2003-2011 and the sample of 9 insurance companies that were operating.

Descriptive statistics and regression analysis were performed to describe the profitability of

insurance companies among insurance companies.

This chapter presents a conclusion of the study by summarizing the study‟s findings and

discussing their implications, and providing suggestions for future research.

The study investigates the impact of firm level characteristics on performance of the

insurance sector of Ethiopia over the period of nine years from 2003 to 2011. For this

purpose, size, volume of capital, age, leverage, liquidity, growth and tangibility are selected

as explanatory variables while ROA is taken as dependent variable. The results of regression

analysis reveal that leverage, size, volume of capital, growth and liquidity are most important

determinant of performance of life insurance sector whereas ROA has statistically

insignificant relationship with, age and tangibility.

5.2 Implications of the Results

The adjusted value of R square (0.43) indicates that performance of insurance companies

is nearly 43% dependent on independent variables i.e. size, leverage, growth, volume of

capital and liquidity. Therefore, it implies that internal factors are important determinants

of profitability of insurance companies in Ethiopia to the extent on average 43% of the

62

change in profitability of the companies can be explained by the selected internal

characteristics.

Negative coefficient of variable liquidity specifies the negative relationship. However,

the relationship between performance and liquidity is statistically significant. Hence,

insurance companies having more liquid assets should find any available investment

alternative. As the findings shows that liquidity and do have negative impact on

profitability, and it provides further implication on the effective risk management

practices in the companies.

The coefficient of variable size is positive and statistically significant at 1% level. This

predicts that performance of large size insurance companies are better than small size

companies. The positive relationship between size and ROA implies that size is used to

capture the fact that larger insurance companies are better placed than smaller once in

harnessing economies of scale in transactions and enjoy a higher level of profits.

The beta values of explanatory variables tangibility and age are with a positive

coefficient sign. However, tangibility and liquidity are not statistically significant with

the large p-values. Therefore, tangibility and age are not considered as powerful

explanatory variables to define the performance of insurance companies in Ethiopia over

nine years.

Leverage is negatively and significantly related with the performance of the insurance

companies. This predicts that the performances of highly levered Ethiopian insurance

companies are going to be less profitable and implies equity financing is better than debt

financing in Ethiopian insurance companies. The leverage ratio level of the insurance

companies affects their profitability negatively, which supports the hypothesis formulated

63

for the study. Thus, from the result it is implied that highly profitable insurance

companies are more likely relied on internally generated funds and equity capital than

debt capital as the source of financing.

The positive and significant relationship between volume of capital and profitability of

insurance companies in Ethiopia implies that a sound capital position is able to pursue

business opportunities more effectively and has more time and flexibility to deal with

problems arising from unexpected losses, thus achieving increased profitability. Hence

indicates that well capitalized insurance companies face lower costs of going bankrupt,

which reduces their cost of funding or that they have lower needs for external funding

which results in higher profitability.

The positive and statistical significant relation between growth rate and profitability of

insurance companies in Ethiopia implies that insurance companies with high rate of

growth in terms of their total assets are also in a better position of being profitable.

5.3 Recommendations for future research

This work is an attempt to study the internal factors affecting profitability of the insurance

sector in Ethiopia. Given the key role that the sector plays in the economy of the country,

future research should focus on both internal and external factors that would provide better

insights for both management and regulatory bodies. Other issues that could be covered in

future research include whether insurance companies effectively and efficiently indemnify

risks and intermediate savings for the provision of risk to the other sectors in the economy,

or whether they allocate resources and manage risks efficiently hence factors affecting

profitability of insurance companies and their implications in risk management practices.

These are important considerations for insurance development in Ethiopi

1

Bibliography

Abreu, M. and Mendes B. (2002), Commercial Bank Interest Margins and Profitability:

Evidence form E.U Countries, Porto Working Paper Series.

Aburime T. (2008), Determinant of Bank Profitability: Company Level Evidence from

Nigeria, Working Paper, Department of Banking and Finance, University of Nigeria.

Adams M., Hardwick P. And Zou H., (2008), “Reinsurance and Corporate Taxation in the

United Kingdom Life Insurance Industry”, Journal of Banking and Finance, Vol. 32.

Agiobenebo, T.J and Ezirim B.C. (2002) “Impact of Financial Intermediation on the

Profitability of Insurance Companies in Nigeria”, First Bank of Nigeria Quarterly

Review; Vol. 2, No.1.

Athanasoglous P. Et Al., (2005), Bank-Specific, Industry- Specific and Macroeconomic

Determinants of Bank Profitability, Bank of Greece Working Paper, No. 25

Belayneh Hailegeorgis Damena (2011), Determinants of Commercial Banks Profitability: An

Empirical Study on Ethiopian Commercial Banks, Addis Ababa University.

Born H. P., (2001), “Insurer Profitability in Different Regulatory and Legal Environments”,

Journal of Regulatory Economics, Vol. 19, Issue. 3.

Demirguc-Kunt, A. And H. Huizinga. (2000). Financial Structure and Bank Profitability.

Desheng Wu Z., Sandra V. &Lianga (2007), “Simultaneous Analysis of Production and

Investment Performance of Canadian Life and Health Insurance Companies”,

Computers and Operations Research, Vol. 34, Issue. 1.

Ephraim W. T. Chirwa (2001) Market Structure, Liberalizationand Performance in the

Malawianbanking Industry, African Economic Research Consortium, Nairobi.

Gujarati, D.N. (1995), Basic Econometrics, Mcgraw-Hill, New York

2

Hailu Zeleke (2007). Insurance in Ethiopia: Historical Development, Present Status and

Future Challenges. Addis Ababa: Master Printing Press.

Hifza Malik, (2011) Determinants of Insurance Companies Profitability: An Analysis of

Insurance Sector Of Pakistan, Academic Research International, Volume 1, Issue 3,

Available at Www.Journals.Savap.Org.Pk.

Kosmidou, K., Pasiouras, F., Zopounidis, C. & Doumpos, M. (2006). A Multivariate

Analysis of the Financial Characteristics of Foreign and Domestic Banks in the UK.

Omega, Vol. 34 ,No. 2.

Kozak S., (2010). Consolidation and Efficiency of the Non-Life Insurance Sector in Poland,

Electronic Journal of Polish Agricultural Universities Vol. 13, Issue 1, Available On:

Http://Www.Ejpau.Media.Pl/Volume13/Issue1/Art-03.Html

Kozak S., (2011) Determinants of Profitability of Non-Life Insurance Companies in Poland

during Integration with the European Financial System, Volume 14, Issue 1.

Available Online: Http://Www.EJPAU.Media.Pl.

Kozo Kiyota, Barbara Peitsch and Robert M Stern (2007), Research Seminar in International

Economics, the Case for Financial Sector Liberalization in Ethiopia, the University of

Michigan.

Levine, R., Loayza, N., & Beck, T. (2000), Financial Intermediation and Growth: Causality

and Causes. Journal of Monetary Economics.

Mattoo, A., Rathindran, R., & Subramanian, A. (2006) Measuring Services Trade

Liberalization and Its Impact on Economic Growth: An Illustration. Journal of

Economic Integration.

3

N.K Childambaran Et.Al. (1997) an Investigation of the Performance of the U.S

Property/Liability Insurance Industry, the Journal of Risk and Insurance, Volume 64.

Naveed Ahmed, Zulfqar Ahmed, Ahmad Usman (2011), Determinants of Performance: A

Case Of Life Insurance Sector of Pakistan, International Research Journal of Finance

and Economics, Eurojournals Publishing, Inc. 2011 Available at

Http://Www.Eurojournals.Com/Finance.Htm

Philip Harwick and Mike Adams (1999) Firm Size and Growth in the UK Insurance Industry.

Rashiah Devinaga (2010) Review of Literature and Theories on Commercial Bank

Profitability, Journal Of Performance Management.

Rebao Chen and Ann Wong (2004) Determinants of Financial Health of Asian Insurance

Companies, the Journal of Risk and Insurance Volume 71 Number 3.

Swiss Re (2008), Profitability of Non-Life Insurance Industry, Egypt, Available At

Www.Insureegypt.Com.

Tesfaye Boru Lelissa (2007), The Impact Of Financial Liberalization On The Ownership,

Market Structure And Performance of The Ethiopian Banking Industry, Addis Ababa

University.

Valentina Flaminin. Et.Al,. (2009), Determinants of Commercial Bank Profitability In Sub-

Saharan Africa, International Monitory Fund (IMF) Working Paper, Wp/09/15.

W.Greene (2000), “Econometric Analysis” Fourth Edition, Prentice Hall.

William H. Greene and Dan Segal (2004) Profitability and Efficiency in the U.S Life

Insurance Industry, Journal of Productivity Analysis, Kluwer Academic Publisher,

Netherlands.

www.nbe.gog.et

4

Wright, K. M. (1992), “The Life Insurance Industry in the United States an Analysis of

Economic and Regulatory Issues”, Country Economics Department the World Bank

Policy Research Working Paper 857.

Yuqi Li (2007) Determinants of Banks‟ Profitability and Its Implication on Risk

Management Practices: Panel Evidence from the UK, the University of Nottingham.

5

Appendices

Appendix 1 Profitability of selected insurance companies (ROA)

AIC AWIC EIC GIC NIC NICE NLIC NYIC UNIC

2002 0.080275 0.134543 0.072501 0.11427 -0.00047 0.035684 0.111769 0.132854 0.068388

2003 0.010684 0.077222 0.071182 0.030963 0.069356 0.047012 0.095333 0.097525 0.074011

2004 -0.00361 0.081567 0.078767 0.033329 -0.10886 0.024431 0.014806 0.082086 -0.04871

2005 0.069724 0.077288 0.082114 0.040445 0.091934 -0.04715 0.039079 0.090256 -0.01355

2006 0.019081 0.056011 0.070755 0.04319 0.046689 0.062004 0.034122 0.093177 0.091099

2007 0.034645 0.044162 0.07209 0.054427 0.075693 0.084913 0.021518 0.092597 0.0896

2008 0.043351 0.062898 0.072601 0.013966 0.103427 0.057178 -0.01439 0.063087 0.154931

2009 0.052638 0.047047 0.077455 0.054114 0.098719 0.046253 0.01973 0.121474 0.041224

2010 0.04799 0.099724 0.092721 0.080492 0.093889 0.058833 0.126543 0.121801 0.130664

2010 0.080275 0.072637 0.087805 0.036416 0.084496 0.002904 0.0902 0.137996 0.076476

6

Appendix 2 Hausman Test for panel regression

Correlated Random Effects - Hausman Test

Equation: EQUATION4

Test cross-section and period random effects

Test Summary Chi-Sq. Statistic Chi-Sq. d.f. Prob.

Cross-section random 2.455350 8 0.9638

Period random 0.538037 8 0.9998

Cross-section and period random 2.909185 7 0.8933

7

Appendix 3: Random Effects Regression out put using E-views

Dependent Variable: ROA

Method: Panel EGLS (Two-way random effects)

Date: 05/22/12 Time: 02:07

Sample: 2003 2011

Periods included: 9

Cross-sections included: 9

Total panel (balanced) observations: 81

Swamy and Arora estimator of component variances

Variable Coefficient Std. Error t-Statistic Prob.

C -0.394827 0.115616 -3.414996 0.0010

LAG 0.181339 0.090731 1.998654 0.0494

AGE 0.001047 0.000861 1.215062 0.2283

SIZE 0.026134 0.006640 3.935510 0.0002

LEV -0.030357 0.005890 -5.154297 0.0000

GR 0.103884 0.021890 4.745643 0.0000

TA 0.067291 0.047764 1.408838 0.1631

LQ -0.011693 0.006650 -1.758351 0.0829 Weighted Statistics

R-squared 0.487931 Mean dependent var 0.060254

Adjusted R-squared 0.438829 S.D. dependent var 0.042579

S.E. of regression 0.031897 Sum squared resid 0.074270

F-statistic 9.936997 Durbin-Watson stat 1.894614

Prob(F-statistic) 0.000000

8

Appendix 4: Table designed for collecting raw Panel financial data to be used in

regression analysis.

Company‟s Name: ____________________________________________

Year of Establishment: _____________________________________

Fisc

al

Year

Amount in ETB

Total Current

Asset

Total Fixed

Asset

Total

asset

Short- term

liability

Long-term

liability

Total

liability

Total

capital

NIBT

2002

2003

2004

2005

2006

2007

2008

2009

2010

2011

9

Appendix 5: Descriptive statistics

ROA LAG AGE SIZE LEV GR VOC TA LQ

Mean 0.060254 0.062355 13.66667 18.72087 2.127482 0.181054 17.62280 0.134005 1.968611

Median 0.069724 0.069724 12.00000 18.70537 1.918916 0.150799 17.69742 0.102243 1.793605

Maximum 0.154931 0.154931 36.00000 21.21965 5.038109 1.102467 19.67227 0.368079 4.300361

Minimum -0.108859 -0.108859 1.000000 16.52656 0.502881 -0.109607 15.81979 0.021042 1.133468

Std. Dev. 0.042579 0.044555 7.439758 1.056107 0.935546 0.180910 0.879214 0.088051 0.691396

Skewness -1.010069 -0.884845 1.498568 0.225803 0.928510 1.972106 0.196677 0.833320 1.804512

Kurtosis 5.469056 4.953500 5.086159 2.881060 3.724736 10.06265 3.018435 2.649716 6.211216

JB 34.34801 23.44938 45.00522 0.736071 13.41145 220.8529 0.523353 9.788819 78.76224

Probability 0.000000 0.000008 0.000000 0.692093 0.001224 0.000000 0.769760 0.007488 0.000000

Sum 4.880583 5.050765 1107.000 1516.391 172.3261 14.66534 1427.447 10.85439 159.4575

Sum Sq. 0.145039 0.158814 4428.000 89.22898 70.01967 2.618270 61.84134 0.620244 38.24224

N 81 81 81 81 81 81 81 81 81

10

Appendix 6: Correlation statistics

Covariance Analysis: Ordinary

Date: 05/13/12 Time: 11:44

Sample: 2003 2011

Included observations: 81

Correlation

t-Statistic

Probability

Observati

ons ROA LAG AGE SIZE LEV GR VOC TA LQ

ROA 1.000000

-----

-----

81

LAG 0.259071* 1.000000

2.384070 -----

0.0195 -----

81 81

AGE 0.222132 0.155454 1.000000

2.024941 1.398705 -----

0.1463 0.1658 -----

81 81 81

SIZE 0.356744** 0.254125** 0.478159* 1.000000

3.394139 2.335377 11.01225 -----

0.0011 0.0221 0.0000 -----

81 81 81 81

LEV -0.044616* -0.038359 0.575873* 0.672534* 1.000000

-0.396952 -0.341189 6.260835 8.077128 -----

0.0025 0.7339 0.0000 0.0000 -----

81 81 81 81 81

GR 0.257726** -0.285424* -

0.19501*** -0.054147 0.076373 1.000000

2.370810 -2.647018 -1.767241 -0.481976 0.680810 -----

0.0202 0.0098 0.0810 0.6312 0.4980 -----

81 81 81 81 81 81

VOC 0.436393* 0.316130* 0.743681* 0.971563* 0.484409* -0.093431 1.000000

4.310882 2.961717 9.887254 36.47021 4.921488 -0.834080 -----

0.0000 0.0040 0.0000 0.0000 0.0000 0.4068 -----

81 81 81 81 81 81 81

TA 0.090525 -0.032067 -0.092996 -0.182436 -0.033366 -0.027489 -0.217501 1.000000

0.807921 -0.285161 -0.830165 -1.649203 -0.296728 -0.244424

-

1.980604** -----

0.4216 0.7763 0.4089 0.1031 0.7675 0.8075 0.0511 -----

81 81 81 81 81 81 81 81

LQ -0.05683*** 0.099973 0.060831 -0.083522 -0.312444* 0.020878 0.023037 -0.435031* 1.000000

-0.505969 0.893051 0.541683 -0.744961 -2.923422 0.185604 0.204808 -4.294284 -----

0.06143 0.3745 0.5896 0.4585 0.0045 0.8532 0.8382 0.0000 -----

81 81 81 81 81 81 81 81 81

*Correlation is significant at 1% level

** Correlation is significant at 5% level

*** Correlation is significant at 10% level

11

Appendix 7: Panel unit root test on ROA

Panel unit root test: Summary

Series: ROA

Date: 05/14/12 Time: 07:05

Sample: 2003 2011

Exogenous variables: Individual effects

User specified lags at: 1

Newey-West bandwidth selection using Bartlett kernel

Balanced observations for each test

Cross-

Method Statistic Prob.** sections

Null: Unit root (assumes common unit root process)

Levin, Lin & Chu t* -7.36522 0.0000 9